Don’t Bet the Farm(ville) on Social Search

By Bryson Meunier, Natural Search Associate Director, Content Solutions

I’m sorry to tell you this way, but if you’re my Facebook friend and you play Farmville or Mafia Wars, I have hidden your status updates about your virtual growing and racketeering.

Same thing applies to my Facebook Foursquare aficionados. If you’ve checked in somewhere recently or become the mayor of something, I have no idea. And I’m fine with that. Don’t get me wrong, I think location sharing is an interesting concept, and there may be applications where I would find the information useful, but my experience has not yet made me aware of these applications.

My Facebook friends and acquaintances are a lively, interesting bunch, and I’m happy to be connected to all of them. Some are connected by blood, some through history and experience, and some through shared professional interests. But I can’t think of one of them that likes every movie, album, book or brand I like, every time, and all the time. In fact, some of my friends’ (actual and Facebook) tastes in things baffle me, and I take their recommendations with a grain of salt if I take them at all.

And when it comes to finding reliable information on some esoteric topic like [search query taxonomy] (an actual search I did today), I might have one or two people in my social circle who even know what I’m talking about, and they’re likely not interested enough to give me good information.

This is why, at this point, I don’t use Facebook as a search engine, and why I’m not really excited about the prospects of social search. I get it in theory: we all have friends whose musical or literary tastes we trust—that hipster or chick who always knows what products to buy or what movies to see, even if no one else is talking about them; and it would be nice to use these trusted preferences as a quality signal to temper the anonymous public opinions that exist on the web. It personalizes things, and therefore makes them more relevant. The theory, I think, is sound.
But in reality there’s Farmville. And Mafia Wars. And The Twilight Saga. And numerous things that even my cool Facebook friends are into that for me come through as noise.

A good social search engine, if there is such a thing, will not just recognize how many of my Facebook friends or Twitter followers are fans of something or discussing a brand, but will know why my friends liked the things that they like and how that’s relevant to me and my query. Something similar to what Google is doing, perhaps, with social search, combining information and opinions from my social circle with other relevant information. Or maybe something like Hunch, Netflix or Amazon, that tries to get to know yours and others’ tastes first before attempting to give you relevant information. But I’m not sure that’s even possible given the recent privacy backlash over Facebook’s open graph.

I’d like to be bullish, and join the crowd of technology consultants using Facebook’s new Open Graph search and “Google killer” in the same headline, but for the most part I think it’s going to be hard for any technology to successfully use social data as a relevance signal, just because so much of it is often irrelevant to the task at hand. No offense to my friends, but if I wanted your opinion I would call you and ask.

As a marketer it’s easy to recognize the opportunity to put your brand in front of 500 million engaged users, and we regularly recommend that brands incorporate social plugins and have some basic strategy for making their brand visible to these growing communities. There is value in these social networks for brands outside of social search.

But let’s not get too excited about social search just yet. Facebook is not yet a Google killer, even with this new announcement. Neither is Twitter. And until they can figure out why millions of us have hidden Farmville and reflect that in the results set, they won’t be any time soon.


Dave McAnally said...

Agreed. Describing Facebook as a Google Killer is sort of like describing your microwave as a cat-dryer. Just because it could do it doesn't mean it'll be particularly good at it, and it certainly isn't what it was really built for in the first place.

That said I think the Open Graph really is a significant milestone unto itself. It doesn't have to be a Google killer to be significant (much as some pundits seem to think otherwise). The ability to harness people's social sphere and turn the casual customer into a brand evangelist has never been easier or more effective. And for end-user, I think it falls to them to decipher the noise from the signal. But that's really always been the case hasn't it? :-)

Copyright © 2008 Resolution Media, Inc. All rights reserved.